Eric Chuk

Personal information

First name

Eric

Last name

Chuk

Institutional affiliation

Department of Information Studies
University of California, Los Angeles
http://is.gseis.ucla.edu/

Narratological interest

I study the logic of events and imagined conditions as expressed in real-world multimedia data. This has also been termed “story logic” or “the logic of intervention.” My goal is to understand why certain events are more open to imagination and how information systems represent events.

My hypothesis is that imaginative language and imagery (such as "if"
statements and computerized animations) are generated by events deemed newsworthy or highly tellable. My approach is to develop a taxonomy of imaginative expression as well as inference rules for newsworthiness by mining an archive of television newscasts, guided by narrative theory and current semantic web standards.

We live in at least two worlds simultaneously: the real or "known" world and the world of imagination. The former is characterized by empiricism, manifested existence, and literal facts, the latter by speculation, the possibility of occurrence, and stories (factual or fictional representations of events, or changes of state). While many information systems have been developed to deal with the real world, far fewer have sought to tap into the world of imagination, which is a rich and complex source of information linked to human desires and emotion.

Currently, the two fields of narratology and ontology for information systems have little bearing on each other despite both addressing the problem of representing events in various contexts, although in distinctly different ways. While narratology has provided numerous theories about what qualifies as an event or what makes something eventful, its scholars have not attempted to apply their insights to the realm beyond literature. Similarly, ontologists seem unaware that storytelling is one of the oldest and arguably most effective means of structuring information, although they have made great strides in codifying their own phenomena of interest for machine processing, mainly in hierarchical and mereological terms.

I argue that a focus on possible events is the key to connecting these two worlds, which should not remain mutually exclusive. Possible events are a special kind of data about the causal relations between certain changes of state, as well as their desirability. So, the indicators of and parameters for possible events should be explored in real world data rather than just literature. The hope is that this move will make information systems, which rarely process imaginative content, more human-oriented by showing how an ontology for possible events might be created inductively. Since ontologies guide our design and usage of systems for navigating information, such an endeavor should build from existing means for representing event data while allowing for this additional layer of semantic complexity, which is provided by narrative.